Jan 2021 – Jan 2022 · 1 yr · Boston, Massachusetts, United States

• Led an organization of 40+ machine learning and software engineers across multiple locations.
• My team was responsible for developing state-of-the-art models to interpret gigapixel pathology images serving a large variety of biopharma clients and use-cases in the translational research and clinical trials space. My team was also responsible for the MLOps infrastructure necessary for workflow orchestration, data sampling, and serving on large models processing images of size 100,000 x 100,000 pixels.

Jan 2017 – Jan 2021 · 4 yrs · Greater New York City Area

• Butterfly developed the most popular hand-held Ultrasound product on the market, capable of scanning any part of the body with a single device. I led a team of machine learning and software engineers that developed and launched a suite of AI products performing real-time clinical measurements and assisting novice ultrasound users to acquire and interpret a broad range of ultrasound views. Launched AI products, several of which obtained FDA clearance (510K) and CE marking, are actively being used by clinicians around the world and include:
• Auto Bladder Volume
• Auto B-Line Counter
• Auto Ejection Fraction
• ScanLab (Educational View Guidance)
• My team also developed tools for dataset creation and visualization, rapid annotation, automated model training, active learning, and monitoring.

Jan 2015 – Jan 2017 · 2 yrs · Greater New York City Area

• Worked in Machine Perception on a variety of projects. Co-created TF-Slim (https://github.com/google-research/tf-slim) which became the basis for the TensorFlow 1.0 API. Co-created gin-config (https://github.com/google/gin-config) which is used across Google Research. Developed and published state-of-the-art approaches for Domain Adaptation and depth prediction. Also contributed to Google Brain's Diabetic Retinopathy and Pathology efforts.

Jun 2014 – Sep 2014 · 3 mos · Mountain View

Jan 2013 – Jan 2013 · 0 mo · Redmond

Sep 2009 – Jan 2015 · 5 yrs 4 mos · New York City

Jun 2008 – Jun 2009 · 1 yr · New York City

• Software Engineer, Prediction API
• Implemented MapReduce-based framework for performing distributed, concurrent classification across multiple classifiers and parameters.
• Software Engineer, Tech Lead, Google’s DoubleClick For Publishers
• Lead 3-person team in developing Google’s DFP SOAP-based public API
• Designed SQL-like Query Language for public API use

Jan 2006 – Jun 2008 · 2 yrs 5 mos · New York City

• Software Engineer, DoubleClick AdExchange
• Designed and implemented public API for AdExchange, DoubleClick's online marketplace for advertising
• Software Engineer, Dart Search
• Developed algorithms for predicting the minimum bids necessary for achieving advertisers' desired positions via Google AdWords, Microsoft AdCenter and Yahoo Advertising

Aug 2004 – Dec 2005 · 1 yr 4 mos · New York City

• Software Engineer, Global Strategic Projects Group
• Conceived of, developed and managed the implementation of a dashboard tool for production support metrics that aggregated data from American, European and Asian regions.